RELATED ART STATEMENT
1,5-AG is a compound which is present in the cerebrospinal fluid and plasma of humans. It is reported that its quantity is markedly reduced in plasma collected from patients with certain diseases, particularly with diabetes (Yasuo Akanuma, Kazuyuki Tobe: Journal of Japanese Internal Medicine Association, 80, 1198-1204, 1991). 1,5-AG is thus expected to be as a marker for diagnosis of diabetes.
As assay for 1,5-AG, there are hitherto known a method based on gas chromatography (Yoshioka, Diabetes, 25, 1115-1118, 1982; hereafter referred to as "GC method") and methods using enzymes (hereafter referred to as "enzymatic methods") such as pyranose oxidase (hereafter abbreviated as PROD) or L-sorbose oxidase (Japanese Patent KOKAI (Laid-Open) No. 63-185397).
Serum or plasma collected from the patient with diabetes is mainly a specimen to be assayed for 1,5-AG. In blood from the patient with diabetes, its glucose concentration is higher than that of normal person. In blood from normal person, the glucose concentration is in the range of approximately 60 to 100 mg/dl, whereas in blood from the patient with diabetes, the glucose concentration is widely distributed in the range of 100 to 1000 mg/dl. On the other hand, the concentration of 1,5-AG in blood is in the range of 1.64 to 2.68 mg/dl for normal person but in the patient with diabetes its concentration is as extremely low as 0.18 to 0.21 mg/dl (Japanese Clinic, 47, 1989, extra issue, Immunological Inspection in Blood and Urinary Chemical Test over Wide Range; first volume, 439-442, Kawai). Therefore, the concentration of 1,5-AG in blood from the patient with diabetes becomes about 1/470 or less. In addition, glucose is structurally similar to 1,5-AG so that it is impossible to perform selective assay in the presence of 1,5-AG and glucose on the current technical level. It is thus essentially required to selectively remove glucose or pretreat specimen by adequately modifying the specimen.
In the GC method, the pretreatment requires removal of glucose and labeling of 1,5-AG which makes procedures complicate and involves analysis over long periods of time. For these reasons, it is difficult to assay a large number of specimens by the GC method. There are thus problems for applying the method to clinical assay.
In the enzymatic method, the pretreatment is performed by removing glucose using an ion exchange column or by modifying glucose through phosphorylation. The enzymatic method is accompanied by considerably complicated separation procedures when glucose is removed using an ion exchange column. Turning to glucose modification by phosphorylation, the optimum reaction conditions for phosphorylation including difference in the optimum pH were different from the optimum conditions for reactions of quantitative assay for 1,5-AG. Therefore, phosphorylation and assay for 1,5-AG must be carried out under different reaction conditions, respectively. In addition, adenosine-5'-triphosphate (hereafter referred to as "ATP") used for phosphorylation has an inhibitory action against PROD. In view of concentration, there is a limit in adding ATP to the assay system for accelerating phosphorylation and hence, it was difficult to terminate phosphorylation rapidly. In any event, it is impossible to perform quantitative assay rapidly by the prior art methods. In particular, any of the prior art methods has not come to be applied to an automated analysis device widely used for various clinical tests.